language model 0.00314899
parsing model 0.002741706
model training 0.0026646350000000003
joint model 0.0025138820000000003
model prediction 0.002403003
prediction model 0.002403003
general model 0.002310093
crf model 0.0021897360000000003
entropy model 0.002185812
linear model 0.002184964
tion model 0.002153266
upp model 0.002152003
training data 0.002133565
fore model 0.002097019
pipeline model 0.002097019
test data 0.0018906970000000002
model 0.00186603
data set 0.001809745
speech data 0.001775394
different training 0.001772919
english data 0.001737541
word error 0.001712832
future word 0.0016933590000000002
treebank data 0.0016865320000000001
current word 0.001665574
language processing 0.001661966
data sets 0.0016021820000000002
top data 0.00158107
event language 0.001576258
joint models 0.001567362
former data 0.001566097
conversional data 0.001565641
data mismatch 0.001565641
iwslt data 0.001565641
data noises 0.001565641
natural language 0.001565047
speech words 0.001562664
rent word 0.001552171
auxiliary word 0.001547751
dependency parsing 0.0014777940000000002
parsing results 0.001461751
prediction models 0.0014564830000000002
preceding words 0.0014130000000000002
parsing tree 0.001401828
streaming words 0.001357155
unfinished words 0.001354457
punctuation prediction 0.001304969
language 0.00128296
different computation 0.001279888
single parsing 0.001273407
training set 0.00127339
segmentation features 0.001260976
various models 0.001254236
joint pos 0.001232243
parsing trees 0.001225818
different outputs 0.00122241
partial parsing 0.001222196
text table 0.001221032
work sentence 0.001217611
upp models 0.001205483
local features 0.001194969
syntactic features 0.001184549
parsing algorithms 0.001181018
other languages 0.001176198
optimal parsing 0.0011589740000000001
prediction performance 0.001157684
parsing process 0.001155552
syntactic information 0.00113045
pendency parsing 0.001127539
prediction results 0.0011230480000000002
punctuation predictions 0.00112231
words 0.00112223
ency parsing 0.001115279
pcfg parser 0.001105203
various punctuation 0.001102722
text figure 0.001098197
ical features 0.001094894
speech text 0.001092204
tactic features 0.001091739
ing performance 0.001087651
sodic features 0.001080094
punctuation knowledge 0.001069643
pos tag 0.001060703
future work 0.001056452
ing results 0.001053015
optimal punctuation 0.001051294
prediction task 0.001051238
previous work 0.001047817
punctuation types 0.001042873
punctuation symbols 0.001041029
the training 0.001040979
baseline method 0.00104057
information extraction 0.001038768
punctuation symbol 0.001038451
new set 0.001037999
ation information 0.001035338
training instances 0.0010323139999999999
text processing 0.001030776
test set 0.001030522
evaluation results 0.00102643
